Power Mutation Particle Swarm Optimization with Hybrid Discrete Variables and its Application to Gear Reducer

Particle Swarm Optimization (PSO) has shown its fast search speed and good search ability in many optimization problems. However, PSO easily suffers from local minima when dealing with complex problems. To enhance the basic PSO, this paper presents an improved PSO algorithm namely PMPSO, which employs a power mutation (PM) on the global particle. It is to hope that the mutation could help particles jump out local optima. Based on Matlab software, Power Mutation Particle Swarm Optimization (PMPSO) algorithm program PMPSO1.0 with hybrid discrete variables was developed. The updating strategy based on power mutation makes the particles of PMPSO maintain the diversity during the iterative process, thus overcomes the defect of premature convergence. Example of gear reducer indicates that compared with the exiting algorithms, PMPSO gets the best result, thus certify the improvement of the algorithm's searching ability by power mutation.